Our Klim-QML project has secured a contractor for the development and implementation of a prototype climate model based on quantum machine learning. A consortium of the consultancy D-fine and the Garching-based neutral atom start-up Planqc will support the DLR team led by Mierk Schwabe and Veronika Eyring from the DLR Institute of Atmospheric Physics in unlocking the computational potential of DLR’s QCI quantum computers for applications in climate research. Planqc is already involved in the DLR QCI project DiNAQC.
With Klim-QML and the contract for D-Fine and Planqc, DLR is further expanding its expertise in quantum machine learning and creating a highly relevant use case. In the future, quantum machine learning should contribute to more accurate and faster climate models and thus improved technology assessments and mitigation recommendations.
Making better models for a riskier future
Human-induced climate change poses enormous challenges to our societal systems. Better climate models can help us better prepare for these risks. Since the robustness of such technology assessments and mitigation recommendations directly depends on the accuracy and speed of climate models, we cannot rely solely on performance improvements in classical supercomputing in the future. That is why the DLR Institute of Atmospheric Physics is exploring how quantum computers and quantum machine learning (QML) can contribute to improving and accelerating climate models through Klim-QML.
To this end, D-fine and Planqc will initially collaborate as a consortium on the development of a concept for the QML climate model and implement the developed algorithms on the quantum computers. As a main task, the companies will support the concrete implementation of the improvement of parameterisations and tuning with quantum computing algorithms.
With the collaboration of D-fine, Planqc and Klim-QML, we are not only expanding our own competencies, but also supporting the synergies with our other QML projects. Above all, however, we are linking the two important topics of quantum computing and climate research and thus want to contribute to a better understanding of the consequences of climate change.
Parameter tuning with quantum computers
The contract with D-fine and Planqc comprises two main work packages: The first deals with the tuning of climate models, in which the free parameters of the model are set correctly. Here, D-fine and Planqc will help with the implementation of tuning using quantum computers, first on a quantum computing simulator and later on a DLR QCI quantum computer; the first of these will be a Toy model.
In a second work package, individual parameterisations based on high-resolution training data will be replaced. Parameterisations in climate models describe processes such as cloud formation that cannot be directly resolved due to limited computing capacity, and contribute significantly to uncertainties in climate projections. The focus is on the parameterisations that are particularly relevant in the Klim-QML use case of ship emissions, i.e. cloud microphysics, cloud cover and turbulence. D-fine will work with Planqc to implement one of these parameterisations using quantum machine learning, first on a simulator and then on a DLR QCI quantum computer.
Verbesserung von Klimamodellen durch Quantum-Machine-Learning
Wir verbessern Klimamodellen mithilfe von Quantum-Machine-Learning für robuste Technologiefolgenabschätzung und Mitigationsempfehlungen.
Cookie Consent with Real Cookie Banner